The use of strategies such as sequential applications and mixtures may increase the control spectrum and the residual effect of chemical control of weed in the cassava crop. Thus, the objective was to evaluate the selectivity and efficacy of sequential applications and tank mixture of herbicides in the control of weed in the cassava crop cultivar 'Baianinha'. The experimental design was of randomized blocks, with eleven treatments and four replicates. The treatments consisted in: harrowed control, control without harrow, clomazone, mesotrione, S-metolachlor, clomazone+S-metolachlor, mesotrione+S-metolachlor, clomazone+mesotrione, sulfentrazone/clomazone, clomazone/[mesotrione+S-metolachlor], S-metolachlor/[mesotrione+clomazone]. The doses used for clomazone, mesotrione, sulfentrazone and S-metolachlor in the single applications, in sequence and in tank mixture were of 1.25, 0.24; 0.6 and 1.92 kg ha -1 , respectively. The first application of the sequential treatments occurred 50 days after planting (DAP) and the second at 92 DAP. The applications of mesotrione, mesotrione+clomazone and mesotrione+S-metolachlor, were carried late, at 92 DAP. Based on the data obtained, it is concluded that the mixture mesotrione+S-metolachlor and the sequential application of clomazone/ [mesotrione+S-metolachlor] were selective to cassava 'Baianinha' and efficient in the weed control (over 80% up to 148 days after planting).